But then I remembered that I'm the guy who's always going on about how running backs are fungible, and how teams shouldn't waste first-round picks on them, blah blah blah. In fact, here are my exact words from a March 11 column:
[I'm not suggesting] teams absolutely, under no circumstances, should [n]ever draft a running back in the first round. But [the numbers] show that comparable first-round production can be found on the cheap. And it doesn't get any cheaper than an undrafted free agent. Now, there are hundreds of counterexamples of second- and third-day afterthoughts who don't make it out of training camp. I'm not recommending teams shouldn't draft a running back no earlier than the fourth round; but as the most fungible position on the roster, it's much easier to replace a 1,000-yard rusher than it is to replace, for example, the left tackle blocking for that 1,000-yard rusher.Hmm, I make a good point. Running backs are easy to find, although you wouldn't know it looking at Pittsburgh's crew. Currently, it's Willie Parker, Najeh Davenport, and Peg-leg Haynes … which explains why the team signed Barlow.
Glancing at Pittsburgh's depth chart inevitably leads to someone asking this question: Tell me again, why the hell didn't the Steelers draft a running back with one of their picks?" I think Mike Tomlin gives the exact right answer: "There were a couple of guys we looked at, but we stayed true to the board and looked at the value of the pick that we had on the board…"
And although some fans might not agree with the draft board (hello, Matt Spaeth!), it's hard to argue with the philosophy. But leading up to the draft, Tomlin said he wanted to add another running back to share Parker's workload. So about that … Najeh and Kevan is the best the team can do?
Two thoughts: First, adding a second back is a splendid idea. I know, running back-by-committee is not earth-shattering stuff, but Football Outsiders has something they call the Theory of 370. Essentially, a running back who rushes 370 or more times in a season, suffers a serious drop in production -- or worse, a serious injury -- the following season. See if these names sound familiar: Jamal Lewis, Jamal Anderson, Shaun Alexander. Yep, they are all victims of 370.
Larry Johnson, who rushed 416 times in 2006, could be next. And he'll have head coach Herm Edwards to thank. Last season, Edwards unapologetically admitted to overworking his star running back, and honestly, because of his hair-brained decision, I wouldn't be surprised if Johnson's leg actually fell off at some point during the upcoming season.
By contrast, Parker only managed 337 carries in '06, but in my mind, that's too many. Cue Davenport and Barlow. I know, I know, neither player is, say, Michael Turner, but this ain't Madden. The idea isn't to stock every position with the best players in the league, the idea is to surround a few franchise players with a really strong supporting cast, and do it all within the structure of the salary cap.
Plus, in real life how practical is it to have two running backs on a roster making starting money? Unlike, say, cornerback, there are diminishing returns to having a stable full of starter-quality runners.
This leads to my second thought: Decent, cheap backup running backs provide more value than their good-but-pricey counterparts. Just look at the Denver Broncos. Yeah, they signed Travis Henry to a big deal this off-season, but he's their No. 1. Behind him are a just a bunch of guys. Guys, by the way, who had some success running the ball in the NFL.
Now we can debate if Barlow even qualifies as decent, but remember, Davenport's already on the team. Barlow will be competing with him for the No. 2 job. Worst case, Davenport keeps his 2006 role. Best case, Barlow thinks it's 2003. But is there any reason to believe Barlow will be anything other than mediocre? Well, no.
Below are Barlow's DPAR and DVOA rankings. In the last column are the Adjusted Line Yards (ALY) rankings. (Remember, DPAR measures a player's total value and DVOA measure a player's value per play. ALY takes all running back carries and assigns responsibility to the offensive line based on yards gained):
Year TEAM DPAR DVOA ALY 2001 SF 14th 10th 2nd 2002 SF 23rd 9th 8th 2003 SF 15th 14th 5th 2004 SF 50th 48th 31st 2005 SF 50th 49th 29th 2006 NYJ 48th 48th 25thNotice anything? Yeah, Barlow had spectacularly awful numbers the last three seasons, but he was also running behind spectacularly awful offensive lines (ALY). But here's the thing: Barlow's teammate in New York, Leon Washington, was 19th in DPAR and 9th in DVOA. And the year before in San Francisco, Frank Gore ranked 23rd in DPAR and 21st in DVOA. To me, this indicates that Washington and Gore are quality running backs (not exactly breaking news, I know) who are able to overcome suspect offensives line while Barlow is who we thought he was.
In one sense that's good; the Steelers know what they're getting. But this also means that Barlow is only as effective as the five fat guys in front of him. He's not making something out of nothing; if Max Starks misses a block -- and we know he will -- forget about it. Which leads me to this: Pittsburgh ranked 22nd in Adjusted Line Yards in 2006. Like Washington and Gore, Parker outperformed his offensive line (14th DPAR, 24th DVOA), but is there any reason to think that Barlow will be anything but average?
The data don't paint a particularly rosy picture. But the Steelers aren't looking for Barlow to carry the ball 250 times. They are looking for someone to spell Parker and play special teams. Right now, Davenport is the No. 2, and if the season started tomorrow, I'd be happy with him in that role. If Barlow beats him out for the job, all the better. If he doesn't, well, it cost the team virtually nothing to find out. And on the upside, that means more John Kuhn for everybody.